Joint ITS- and IRS-Assisted Cell-Free Networks

Abstract

This paper investigates a joint intelligent transmissive surface (ITS) and intelligent reflecting surface (IRS)-assisted cell-free network. Specifically, various ITS-assisted base stations (BSs), handled via a central processing unit (CPU), broadcast information signals to multiple IoT devices, carried by active transmit beamforming and transmissive reflecting phase shifts. Meanwhile, an IRS passively reflect signals from the ITS-assisted BS to the IoT devices. To examine this network performance, a sum rate is maximized among all users to jointly optimize the active beamforming, ITS and IRS passive beampatterns. These coupled variables leads to the non-convexity of this formulated optimization problem, which cannot be directly solved. To deal with this issue, we begin with applying the Lagrange dual transformation (LDT) and quadratic transformation (QT) to recast the sum of multiple logarithmically fractional objectives to the subtractive form, and further to quadratic form. Next, an alternating optimization (AO) algorithm is presented to separately the active beamforming, ITS and IRS passive beampatterns in an iterative fashion. Each sub-optimal solution of these variables can be iteratively derived, in a closed-form, by solving the quadratic objective function with a convex constraint or a unitmodulus constraint via the dual method with bisection search or the Alternating Direction Method of Multipliers (ADMM) algorithm. Finally, simulation results are provided to confirm the performance of the proposed algorithm compared to several benchmark schemes. IndexTerms-Cell-free networks, intelligent reflecting/transmissive surface (IRS/ITS), Lagrange dual transformation (LDT), quadratic transformation (QT), and alternating direction method of multipliers (ADMM).